Built with Michael Ren, this project explored a range of machine learning models to classify traffic signs from images. Our top performer—a CNN—hit 99.8% accuracy on unseen data. All findings captured in a full report.
Winner of the 2023 CISSA Catalyst Hackathon. This AI-powered web app builds personalized Spotify playlists based on your surroundings, mood, and music taste—developed in collaboration with a team for the 48-hour challenge.
A playful interactive visualization built with TensorFlow and UMAP, letting users explore word embeddings in 3D—rotate, zoom, and add words in real time. Created for the DSCubed website with the DSCubed IT team.
A high-level game-playing agent for Freckers—a checkers-style game with chain-jumping frogs. Combines Minimax with alpha-beta pruning and Monte Carlo Tree Search to outplay human opponents. Built from scratch to master strategy and board control.